68 ROYAL SOCIETY OF CANADA 
through, and the heat will be able to penetrate into the subsoil, This 
must have a very marked effect on the vigour of the growth of trees, and 
on the flow of the sap, which is doubtless influenced as much by the tem- 
perature and condition of the soil about the roots as by the prevailing 
atmospheric conditions. 
DIURNAL VARIATIONS OF TEMPERATURE IN THE SURFACE Sotn. 
Observations of the diurnal variations of temperature were taken, 
as soon as the ground began to thaw, at frequent intervals throughout 
the day and night. These are illustrated, for two typical periods of five 
days each, by the curves showing the variations of temperature of each 
of the four upper thermometers. [It may be noticed that the curves are 
somewhat smoother than similar curves obtained with mercury ther- 
mometers. This is probably explained by the fact that the thermometers 
are laid horizontally, and are not disturbed in reading. 
The first set of curves, Fig. 2, illustrates the period of thawing of the 
ground. The waves of the diurnal variation of the 4-inch thermometer 
have an amplitude of between six and eight degrees, with a maximum 
occurring at about 6 pm. The amplitude is greatly affected by the 
clearness of the atmosphere, and would probably afford a convenient 
means of measuring the mean daily intensity of radiation. On April 
20th, for instance, with no material variation in the air temperature, the 
maximum attained was considerably diminished by the presence of small 
cumulus clouds causing occasional obscuration of the sun. Similarly, on 
the night of April 21st the rate of fall, and the form of the curve were 
completely changed by a thin veil of cirrhus clouds, which checked the 
radiation. 
At a depth of 10 inches the amplitude is reduced to about a quarter, 
and the phase still further retarded, the minimum occurring about mid- 
day. 
At a depth of 20 inches the diurnal variation can still be traced, but 
does not usually amount to more than two or three-tenths of a degree F. 
In Fig. 3, the variations of the air and surface temperature are shown, 
together with those of the 4-inch thermometer, for the five extremely hot 
days, May 6th to luth. The observations of the air temperature were 
not sufficiently numerous to give a continuous record, but suffice to indi- 
cate the general nature of the changes. The surface thermometer, which 
had been lying on the top of the grass, beneath the snow-layer, through- 
out the winter, was, for these observations, laid in a horizontal hole bored 
at a depth of 1 inch. 
It will be observed that the maximum surface temperature on clear 
days occurred at about 1 p.m., whereas the maximum air temperature 
occurred at about 3 p.m. This was, perhaps, due in some measure to 
partial shadows cast by trees in the neighbourhood, which fell on the 
